Causal Graph Neural Network Predicts Financial Systemic Risk and Contagion Trajectories
Researchers developed CSTGL, a causal spatio-temporal graph learning architecture that improves banking systemic risk detection and models regulatory counterfactual interventions.

